Article(id=1148106707026309858, tenantId=1146029695717560320, journalId=1146031787341344770, issueId=1148106697601704181, articleNumber=1003-3033(2025)01-0112-08, orderNo=null, doi=10.16265/j.cnki.issn1003-3033.2025.01.0540, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=null, receivedDate=1723305600000, receivedDateStr=2024-08-11, revisedDate=1729353600000, revisedDateStr=2024-10-20, acceptedDate=null, acceptedDateStr=null, onlineDate=1751659569746, onlineDateStr=2025-07-05, pubDate=1737993600000, pubDateStr=2025-01-28, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1751659569746, onlineIssueDateStr=2025-07-05, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1751659569746, creator=13701087609, updateTime=1751659569746, updator=13701087609, issue=Issue{id=1148106697601704181, tenantId=1146029695717560320, journalId=1146031787341344770, year='2025', volume='35', issue='1', pageStart='1', pageEnd='252', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=0, createTime=1751659567499, creator=13701087609, updateTime=1757401533944, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1172190250475573883, tenantId=1146029695717560320, journalId=1146031787341344770, issueId=1148106697601704181, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1172190250475573884, tenantId=1146029695717560320, journalId=1146031787341344770, issueId=1148106697601704181, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=112, endPage=119, ext={EN=ArticleExt(id=1149757469364240876, articleId=1148106707026309858, tenantId=1146029695717560320, journalId=1146031787341344770, language=EN, title=Research on vehicle hazardous cut-in strategy used in autonomous driving test, columnId=1149733269173878863, journalTitle=China Safety Science Journal, columnName=Safety engineering technology, runingTitle=null, highlight=null, articleAbstract=

To improve the interaction ability of traffic vehicles in the cut-in scenario,a method for constructing a vehicle hazardous cut-in strategy based on deep reinforcement learning was proposed. Firstly,a simulated environment was built based on scalable multi-agent reinforcement learning training school(SMARTS) simulation platform. Then,twin delayed deep deterministic policy gradients (TD3) algorithm was adopted to train an agent to cut in a randomly chosen target vehicle hazardously. The algorithm was compared with proximal policy optimization (PPO) and deep deterministic policy gradient (DDPG) algorithms. The trained model was tested in seven different scenarios with varying traffic densities. Finally,a multi-agent testing environment was built,and the trained model was applied to validate intelligent driving strategies. The results show that the success rate of hazardous cut-ins reaches 80.35% in model training with TD3 algorithm,outperforming both comparative methods. In model testing,except for the 2 700 vehicle/h test scenario,the model achieves a hazardous cut-in success rate of over 80% in the other three test scenarios that were not used in training,demonstrating good generalization ability. Meanwhile,the time to collision values between the ego vehicle and the target vehicle at the moment of lane changes are concentrated within the range of 0 to 6 seconds,with 95% falling within this bracket. The proportions of time to collision values in the intervals of (0,2],(2,4],(4,6]s are 60%,30%,and 5% respectively,covering test conditions with different collision risk. In the validation of intelligent driving strategies,the traffic vehicle controlled by the trained model can actively perform cut-ins in front of the test vehicles,exposing it to the risk of a rear-end collision and helping in identifying safety vulnerabilities in intelligent driving strategies.

, correspAuthors=Yunxing CHEN, authorNote=null, correspAuthorsNote=null, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=null, magXml=null, pdfUrl=null, pdf=null, pdfFileSize=null, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=null, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=null, mapNumber=null, authorCompany=null, fund=null, authors=null, authorsList=Yang ZHOU, Yunxing CHEN, Ling WU), CN=ArticleExt(id=1148106714529919394, articleId=1148106707026309858, tenantId=1146029695717560320, journalId=1146031787341344770, language=CN, title=用于自动驾驶测试的车辆危险切入策略研究, columnId=1149733269727526997, journalTitle=中国安全科学学报, columnName=安全工程技术, runingTitle=null, highlight=null, articleAbstract=

为提高车辆切入测试场景中交通车的交互能力,提出一种基于深度强化学习方法的车辆危险切入策略设计方法。首先,基于可扩展多智能体强化学习培训学校(SMARTS)仿真平台构建仿真环境;然后,采用双延迟深度确定性策略梯度算法(TD3)训练智能体危险切入随机选定的目标车辆,将该算法与近端策略优化算法(PPO)和深度确定性策略梯度算法(DDPG)进行对比,在7种不同车辆密度的场景中测试训练后的模型;最后,构建多智能体测试环境,将所训练模型用于智能驾驶策略的验证。结果表明:模型在训练中的危险切入成功率达80.35%,优于2种对比方法;在模型测试中,除2 700辆/h测试场景外,该模型在另外3个未在训练中使用的测试场景均达到80%以上的危险切入成功率,显示出良好的泛化能力。同时,切入时刻与目标车的碰撞时间值显示95%集中在0~6 s,取值在(0,2]、(2,4]和(4,6]s的占比分别为60%、30%和5%,可覆盖具有不同碰撞风险的测试工况。在智能驾驶策略验证中,采用所训练模型控制的交通车能主动切入至待测车辆前方,使待测车辆面临追尾风险,有助于发现智能驾驶策略的安全隐患。

, correspAuthors=陈运星, authorNote=null, correspAuthorsNote=
**陈运星(1987—),男,湖北荆门人,博士,副教授,主要从事驾驶行为感知、智能驾驶技术等方面的研究。E-mail:
, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=2k/R0VnDWCmhP/9trkWbZw==, magXml=d07YNGWoelcCiuo3ZJPEfw==, pdfUrl=null, pdf=UrVqRffUSnDfLmmYabKgxA==, pdfFileSize=null, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=null, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=gKgb924HpOtACpGmPxtRNA==, mapNumber=null, authorCompany=null, fund=null, authors=

周扬 (1989—),男,陕西汉中人,博士,副教授,主要从事人-车-路系统安全、自动驾驶测试等方面的研究。E-mail:

陈运星 副教授

吴玲 副教授

, authorsList=周扬, 陈运星, 吴玲)}, authors=[Author(id=1172170789508989484, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=297399014@qq.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1172170789580292656, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, authorId=1172170789508989484, language=EN, stringName=Yang ZHOU, firstName=Yang, middleName=null, lastName=ZHOU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 School of Vehicle Engineering,Xi'an Aeronautical Institute,Xi'an Shaanxi 710077,China
2 Hubei Key Laboratory of Power System Design and Test for Electrical Vehicle,Hubei University of Arts and Science,Xiangyang Hubei 441053,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1172170789639012913, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, authorId=1172170789508989484, language=CN, stringName=周扬, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 西安航空学院 车辆工程学院,陕西 西安 710077
2 湖北文理学院 纯电动汽车动力系统设计与测试湖北省重点实验室,湖北 襄阳 441053, bio={"img":"ZnkxmKJgKDNzhAePr+ktSQ==","content":"

周扬 (1989—),男,陕西汉中人,博士,副教授,主要从事人-车-路系统安全、自动驾驶测试等方面的研究。E-mail:

"}, bioImg=ZnkxmKJgKDNzhAePr+ktSQ==, bioContent=

周扬 (1989—),男,陕西汉中人,博士,副教授,主要从事人-车-路系统安全、自动驾驶测试等方面的研究。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1172170789265719842, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, xref=1, ext=[AuthorCompanyExt(id=1172170789274108451, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, companyId=1172170789265719842, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 School of Vehicle Engineering,Xi'an Aeronautical Institute,Xi'an Shaanxi 710077,China), AuthorCompanyExt(id=1172170789282497060, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, companyId=1172170789265719842, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 西安航空学院 车辆工程学院,陕西 西安 710077)]), AuthorCompany(id=1172170789349605925, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, xref=2, ext=[AuthorCompanyExt(id=1172170789357994534, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, companyId=1172170789349605925, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 Hubei Key Laboratory of Power System Design and Test for Electrical Vehicle,Hubei University of Arts and Science,Xiangyang Hubei 441053,China), AuthorCompanyExt(id=1172170789366383143, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, companyId=1172170789349605925, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 湖北文理学院 纯电动汽车动力系统设计与测试湖北省重点实验室,湖北 襄阳 441053)])]), Author(id=1172170789701927475, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=chenyunxing@chd.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1172170789769036342, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, authorId=1172170789701927475, language=EN, stringName=Yunxing CHEN, firstName=Yunxing, middleName=null, lastName=CHEN, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, 3, **, address=2 Hubei Key Laboratory of Power System Design and Test for Electrical Vehicle,Hubei University of Arts and Science,Xiangyang Hubei 441053,China
3 School of Automotive and Traffic Engineering,Hubei University of Arts and Science,Xiangyang Hubei 441053,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1172170789823562295, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, authorId=1172170789701927475, language=CN, stringName=陈运星, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, 3, **, address=2 湖北文理学院 纯电动汽车动力系统设计与测试湖北省重点实验室,湖北 襄阳 441053
3 湖北文理学院 汽车与交通工程学院,湖北 襄阳 441053, bio={"content":"

陈运星 副教授

"}, bioImg=null, bioContent=

陈运星 副教授

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1172170789349605925, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, xref=2, ext=[AuthorCompanyExt(id=1172170789357994534, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, companyId=1172170789349605925, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 Hubei Key Laboratory of Power System Design and Test for Electrical Vehicle,Hubei University of Arts and Science,Xiangyang Hubei 441053,China), AuthorCompanyExt(id=1172170789366383143, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, companyId=1172170789349605925, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 湖北文理学院 纯电动汽车动力系统设计与测试湖北省重点实验室,湖北 襄阳 441053)]), AuthorCompany(id=1172170789425103400, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, xref=3, ext=[AuthorCompanyExt(id=1172170789433492009, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, companyId=1172170789425103400, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 School of Automotive and Traffic Engineering,Hubei University of Arts and Science,Xiangyang Hubei 441053,China), AuthorCompanyExt(id=1172170789446074922, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, companyId=1172170789425103400, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 湖北文理学院 汽车与交通工程学院,湖北 襄阳 441053)])]), Author(id=1172170789886476857, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1172170789987140155, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, authorId=1172170789886476857, language=EN, stringName=Ling WU, firstName=Ling, middleName=null, lastName=WU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 School of Vehicle Engineering,Xi'an Aeronautical Institute,Xi'an Shaanxi 710077,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1172170790054249020, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, authorId=1172170789886476857, language=CN, stringName=吴玲, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 西安航空学院 车辆工程学院,陕西 西安 710077, bio={"content":"

吴玲 副教授

"}, bioImg=null, bioContent=

吴玲 副教授

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1172170789265719842, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, xref=1, ext=[AuthorCompanyExt(id=1172170789274108451, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, companyId=1172170789265719842, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 School of Vehicle Engineering,Xi'an Aeronautical Institute,Xi'an Shaanxi 710077,China), AuthorCompanyExt(id=1172170789282497060, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, companyId=1172170789265719842, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 西安航空学院 车辆工程学院,陕西 西安 710077)])])], keywords=[Keyword(id=1172170790222021181, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, language=EN, orderNo=1, keyword=autonomous driving), Keyword(id=1172170790364627518, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, language=EN, orderNo=2, keyword=vehicle hazardous cut-in), Keyword(id=1172170790456902207, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, language=EN, orderNo=3, keyword=virtual tests), Keyword(id=1172170790519816768, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, language=EN, orderNo=4, keyword=hazardous scenarios), Keyword(id=1172170790586925633, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, language=EN, orderNo=5, keyword=reinforcement learning), Keyword(id=1172170790637257282, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, language=CN, orderNo=1, keyword=自动驾驶), Keyword(id=1172170790708560451, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, language=CN, orderNo=2, keyword=车辆危险切入), Keyword(id=1172170790763086404, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, language=CN, orderNo=3, keyword=虚拟测试), Keyword(id=1172170790826000965, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, language=CN, orderNo=4, keyword=危险场景), Keyword(id=1172170790897304134, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, language=CN, orderNo=5, keyword=强化学习)], refs=[Reference(id=1172170793023816289, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, doi=null, pmid=null, pmcid=null, year=2019, volume=32, issue=6, pageStart=1, pageEnd=19, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=朱冰, 张培兴, 赵健, journalName=中国公路学报, refType=null, unstructuredReference=朱冰, 张培兴, 赵健, 等. 基于场景的自动驾驶汽车虚拟测试研究进展[J]. 中国公路学报, 2019, 32(6): 1-19., articleTitle=基于场景的自动驾驶汽车虚拟测试研究进展, refAbstract=null), Reference(id=1172170793099313762, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, doi=null, pmid=null, pmcid=null, year=2019, volume=32, issue=6, pageStart=1, pageEnd=19, url=null, language=null, rfNumber=[1], rfOrder=1, authorNames=ZHU Bing, ZHANG Peixing, ZHAO Jian, journalName=China Journal of Highway and Transport, refType=null, unstructuredReference=ZHU Bing, ZHANG Peixing, ZHAO Jian, et al. Review of scenario-based virtual validation methods for automated vehicle[J]. China Journal of Highway and Transport, 2019, 32(6): 1-19., articleTitle=Review of scenario-based virtual validation methods for automated vehicle, refAbstract=null), Reference(id=1172170793174811235, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, doi=null, pmid=null, pmcid=null, year=2022, volume=35, issue=1, pageStart=316, pageEnd=333, url=null, language=null, rfNumber=[2], rfOrder=2, authorNames=邓伟文, 李江坤, 任秉韬, journalName=中国公路学报, refType=null, unstructuredReference=邓伟文, 李江坤, 任秉韬, 等. 面向自动驾驶的仿真场景自动生成方法综述[J]. 中国公路学报, 2022, 35(1): 316-333., articleTitle=面向自动驾驶的仿真场景自动生成方法综述, refAbstract=null), Reference(id=1172170793250308708, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, doi=null, pmid=null, pmcid=null, year=2022, volume=35, issue=1, pageStart=316, pageEnd=333, url=null, language=null, rfNumber=[2], rfOrder=3, authorNames=DENG Weiwen, LI Jiangkun, REN Bingtao, journalName=China Journal of Highway and Transport, refType=null, unstructuredReference=DENG Weiwen, LI Jiangkun, REN Bingtao, et al. A survey on automatic simulation scenario generation methods for autonomous driving[J]. China Journal of Highway and Transport, 2022, 35(1): 316-333., articleTitle=A survey on automatic simulation scenario generation methods for autonomous driving, refAbstract=null), Reference(id=1172170793342583397, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, doi=null, pmid=null, pmcid=null, year=2021, volume=43, issue=4, pageStart=610, pageEnd=619, url=null, language=null, rfNumber=[3], rfOrder=4, authorNames=徐向阳, 胡文浩, 董红磊, journalName=汽车工程, refType=null, unstructuredReference=徐向阳, 胡文浩, 董红磊, 等. 自动驾驶汽车测试场景构建关键技术综述[J]. 汽车工程, 2021, 43(4): 610-619., articleTitle=自动驾驶汽车测试场景构建关键技术综述, refAbstract=null), Reference(id=1172170793405497958, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, doi=null, pmid=null, pmcid=null, year=2021, volume=43, issue=4, pageStart=610, pageEnd=619, url=null, language=null, rfNumber=[3], rfOrder=5, authorNames=XU Xiangyang, HU Wenhao, DONG Honglei, journalName=Automotive Engineering, refType=null, unstructuredReference=XU Xiangyang, HU Wenhao, DONG Honglei, et al. Review of key technologies for autonomous vehicle test scenario construction[J]. Automotive Engineering, 2021, 43(4): 610-619., articleTitle=Review of key technologies for autonomous vehicle test scenario construction, refAbstract=null), Reference(id=1172170793560687207, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, doi=null, pmid=null, pmcid=null, year=2023, volume=53, issue=11, pageStart=3130, pageEnd=3140, url=null, language=null, rfNumber=[4], rfOrder=6, authorNames=郭柏苍, 雒国凤, 金立生, journalName=吉林大学学报:工学版, refType=null, unstructuredReference=郭柏苍, 雒国凤, 金立生, 等. 面向自动驾驶虚拟测试的变道切入场景库构建方法[J]. 吉林大学学报:工学版, 2023, 53(11): 3130-3140., articleTitle=面向自动驾驶虚拟测试的变道切入场景库构建方法, refAbstract=null), Reference(id=1172170793678127720, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, doi=null, pmid=null, pmcid=null, year=2023, volume=53, issue=11, pageStart=3130, pageEnd=3140, url=null, language=null, rfNumber=[4], rfOrder=7, authorNames=GUO Baicang, LUO Guofeng, JIN Lisheng, journalName=Journal of Jilin University: Engineering and Technology Edition, refType=null, unstructuredReference=GUO Baicang, LUO Guofeng, JIN Lisheng, et al. Construction method of cut-in scenario library for automatic driving virtual tests[J]. Journal of Jilin University: Engineering and Technology Edition, 2023, 53(11): 3130-3140., articleTitle=Construction method of cut-in scenario library for automatic driving virtual tests, refAbstract=null), Reference(id=1172170793766208105, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, doi=null, pmid=null, pmcid=null, year=2023, volume=49, issue=10, pageStart=2211, pageEnd=2223, url=null, language=null, rfNumber=[5], rfOrder=8, authorNames=赵祥模, 赵玉钰, 景首才, journalName=自动化学报, refType=null, unstructuredReference=赵祥模, 赵玉钰, 景首才, 等. 面向自动驾驶测试的危险变道场景泛化生成[J]. 自动化学报, 2023, 49 (10): 2211-2223., articleTitle=面向自动驾驶测试的危险变道场景泛化生成, refAbstract=null), Reference(id=1172170793845899882, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, doi=null, pmid=null, pmcid=null, year=2023, volume=49, issue=10, pageStart=2211, pageEnd=2223, url=null, language=null, rfNumber=[5], rfOrder=9, authorNames=ZHAO Xiangmo, ZHAO Yuyu, JING Shoucai, journalName=Acta Automatica Sinica, refType=null, unstructuredReference=ZHAO Xiangmo, ZHAO Yuyu, JING Shoucai, et al. Generalization generation of hazardous lane-changing scenarios for automated vehicle testing[J]. Acta Automatica Sinica, 2023, 49 (10): 2211-2223., articleTitle=Generalization generation of hazardous lane-changing scenarios for automated vehicle testing, refAbstract=null), Reference(id=1172170793921397355, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, doi=null, pmid=null, pmcid=null, year=2023, volume=53, issue=3, pageStart=704, pageEnd=712, url=null, language=null, rfNumber=[6], rfOrder=10, authorNames=朱冰, 范天昕, 赵健, journalName=吉林大学学报:工学版, refType=null, unstructuredReference=朱冰, 范天昕, 赵健, 等. 基于危险边界搜索的自动驾驶系统加速测试方法[J]. 吉林大学学报:工学版, 2023, 53(3): 704-712., articleTitle=基于危险边界搜索的自动驾驶系统加速测试方法, refAbstract=null), Reference(id=1172170793984311916, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, doi=null, pmid=null, pmcid=null, year=2023, volume=53, issue=3, pageStart=704, pageEnd=712, url=null, language=null, rfNumber=[6], rfOrder=11, authorNames=ZHU Bing, FAN Tianxin, ZHAO Jian, journalName=Journal of Jilin University: Engineering and Technology Edition, refType=null, unstructuredReference=ZHU Bing, FAN Tianxin, ZHAO Jian, et al. Accelerate test method of automated driving system based on hazardous boundary search[J]. Journal of Jilin University: Engineering and Technology Edition, 2023, 53(3): 704-712., articleTitle=Accelerate test method of automated driving system based on hazardous boundary search, refAbstract=null), Reference(id=1172170794047226477, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, doi=null, pmid=null, pmcid=null, year=2021, volume=12, issue=1, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[7], rfOrder=12, authorNames=FENG Shuo, YAN Xintao, SUN Haowei, journalName=Nature Communications, refType=null, unstructuredReference=FENG Shuo, YAN Xintao, SUN Haowei, et al. Intelligent driving intelligence test for autonomous vehicles with naturalistic and adversarial environment[J]. Nature Communications, 2021, 12(1): DOI: 10.1038/s41467-021-21007-8., articleTitle=Intelligent driving intelligence test for autonomous vehicles with naturalistic and adversarial environment, refAbstract=null), Reference(id=1172170794114335342, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, doi=null, pmid=null, pmcid=null, year=2023, volume=615, issue=7953, pageStart=620, pageEnd=627, url=null, language=null, rfNumber=[8], rfOrder=13, authorNames=FENG Shuo, SUN Haowei, YAN Xintao, journalName=Nature, refType=null, unstructuredReference=FENG Shuo, SUN Haowei, YAN Xintao, et al. Dense reinforcement learning for safety validation of autonomous vehicles[J]. Nature, 2023, 615(7953): 620-627., articleTitle=Dense reinforcement learning for safety validation of autonomous vehicles, refAbstract=null), Reference(id=1172170794194027119, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, doi=null, pmid=null, pmcid=null, year=2021, volume=2675, issue=11, pageStart=587, pageEnd=600, url=null, language=null, rfNumber=[9], rfOrder=14, authorNames=SUN Haowei, FENG Shuo, YAN Xintao, journalName=Transportation Research Record, refType=null, unstructuredReference=SUN Haowei, FENG Shuo, YAN Xintao, et al. Corner case generation and analysis for safety assessment of autonomous vehicles[J]. Transportation Research Record, 2021, 2675(11): 587-600., articleTitle=Corner case generation and analysis for safety assessment of autonomous vehicles, refAbstract=null), Reference(id=1172170794261135984, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, doi=null, pmid=null, pmcid=null, year=2022, volume=44, issue=7, pageStart=976, pageEnd=986, url=null, language=null, rfNumber=[10], rfOrder=15, authorNames=李江坤, 邓伟文, 任秉韬, journalName=汽车工程, refType=null, unstructuredReference=李江坤, 邓伟文, 任秉韬, 等. 基于场景动力学和强化学习的自动驾驶边缘测试场景生成方法[J]. 汽车工程, 2022, 44 (7): 976-986., articleTitle=基于场景动力学和强化学习的自动驾驶边缘测试场景生成方法, refAbstract=null), Reference(id=1172170794324050545, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, doi=null, pmid=null, pmcid=null, year=2022, volume=44, issue=7, pageStart=976, pageEnd=986, url=null, language=null, rfNumber=[10], rfOrder=16, authorNames=LI Jiangkun, DENG Weiwen, REN Bingtao, journalName=Automotive Engineering, refType=null, unstructuredReference=LI Jiangkun, DENG Weiwen, REN Bingtao, et al. Automatic driving edge scene generation method based on scene dynamics and reinforcement learning[J]. Automotive Engineering, 2022, 44(7): 976-986., articleTitle=Automatic driving edge scene generation method based on scene dynamics and reinforcement learning, refAbstract=null), Reference(id=1172170794403742322, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=null, pageStart=264, pageEnd=285, url=null, language=null, rfNumber=[11], rfOrder=17, authorNames=ZHOU Ming, LUO Jun, VILLELLA J, journalName=Conference on Robot Learning, refType=null, unstructuredReference=ZHOU Ming, LUO Jun, VILLELLA J, et al. Smarts: an open-source scalable multi-agent reinforcement training school for autonomous driving[C]. Conference on Robot Learning, 2021: 264-285., articleTitle=Smarts: an open-source scalable multi-agent reinforcement training school for autonomous driving, refAbstract=null), Reference(id=1172170794466656883, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, doi=null, pmid=null, pmcid=null, year=2022, volume=32, issue=8, pageStart=140, pageEnd=145, url=null, language=null, rfNumber=[12], rfOrder=18, authorNames=罗崎瑞, 张道文, 周华, journalName=中国安全科学学报, refType=null, unstructuredReference=罗崎瑞, 张道文, 周华, 等. 面向智能汽车预期功能安全的驾驶场景评价[J]. 中国安全科学学报, 2022, 32(8):140-145., articleTitle=面向智能汽车预期功能安全的驾驶场景评价, refAbstract=null), Reference(id=1172170794533765748, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, doi=null, pmid=null, pmcid=null, year=2022, volume=32, issue=8, pageStart=140, pageEnd=145, url=null, language=null, rfNumber=[12], rfOrder=19, authorNames=LUO Qirui, ZHANG Daowen, ZHOU Hua, journalName=China Safety Science Journal, refType=null, unstructuredReference=LUO Qirui, ZHANG Daowen, ZHOU Hua, et al. Evaluation on driving scenarios for safety of intended functionality of intelligent vehicles[J]. China Safety Science Journal, 2022, 32(8):140-145., articleTitle=Evaluation on driving scenarios for safety of intended functionality of intelligent vehicles, refAbstract=null), Reference(id=1172170794609263221, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, doi=null, pmid=null, pmcid=null, year=2023, volume=51, issue=9, pageStart=1424, pageEnd=1432, url=null, language=null, rfNumber=[13], rfOrder=20, authorNames=胡祥旺, 倪颖, 孙剑, journalName=同济大学学报:自然科学版, refType=null, unstructuredReference=胡祥旺, 倪颖, 孙剑. 车联网环境下匝道汇入区瓶颈换道优化[J]. 同济大学学报:自然科学版, 2023, 51(9):1424-1432., articleTitle=车联网环境下匝道汇入区瓶颈换道优化, refAbstract=null), Reference(id=1172170794672177782, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, doi=null, pmid=null, pmcid=null, year=2023, volume=51, issue=9, pageStart=1424, pageEnd=1432, url=null, language=null, rfNumber=[13], rfOrder=21, authorNames=HU Xiangwang, NI Ying, SUN Jian, journalName=Journal of Tongji University: Natural Science, refType=null, unstructuredReference=HU Xiangwang, NI Ying, SUN Jian. Freeway merging area lane changing advisory optimization under connected vehicles environment[J]. Journal of Tongji University: Natural Science, 2023, 51(9): 1424-1432., articleTitle=Freeway merging area lane changing advisory optimization under connected vehicles environment, refAbstract=null), Reference(id=1172170794743480951, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=1, pageStart=11, pageEnd=18, url=null, language=null, rfNumber=[14], rfOrder=22, authorNames=周文帅, 朱宇, 赵祥模, journalName=汽车技术, refType=null, unstructuredReference=周文帅, 朱宇, 赵祥模, 等. 面向高速公路车辆切入场景的自动驾驶测试用例生成方法[J]. 汽车技术, 2021(1): 11-18., articleTitle=面向高速公路车辆切入场景的自动驾驶测试用例生成方法, refAbstract=null), Reference(id=1172170794823172728, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=1, pageStart=11, pageEnd=18, url=null, language=null, rfNumber=[14], rfOrder=23, authorNames=ZHOU Wenshuai, ZHU Yu, ZHAO Xiangmo, journalName=Automobile Technology, refType=null, unstructuredReference=ZHOU Wenshuai, ZHU Yu, ZHAO Xiangmo, et al. Vehicle cut-in test case generation methods for testing of autonomous driving on highway[J]. Automobile Technology, 2021(1): 11-18., articleTitle=Vehicle cut-in test case generation methods for testing of autonomous driving on highway, refAbstract=null), Reference(id=1172170794890281593, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, doi=null, pmid=null, pmcid=null, year=2018, volume=null, issue=null, pageStart=1587, pageEnd=1596, url=null, language=null, rfNumber=[15], rfOrder=24, authorNames=FUJIMOTO S, HOOF H, MEGER D, journalName=International Conference on Machine Learning (ICML), refType=null, unstructuredReference=FUJIMOTO S, HOOF H, MEGER D. Addressing function approximation error in actor-critic methods[C]. International Conference on Machine Learning (ICML), 2018: 1587-1596., articleTitle=Addressing function approximation error in actor-critic methods, refAbstract=null), Reference(id=1172170794953196154, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=2023, pageEnd=null, url=null, language=null, rfNumber=[16], rfOrder=25, authorNames=ZHOU Yang, CHEN Yunxing, journalName=Journal of Advanced Transportation, refType=null, unstructuredReference=ZHOU Yang, CHEN Yunxing. Learning to drive in the NGSIM simulator using proximal policy optimization[J]. Journal of Advanced Transportation, 2023, 2023: DOI: 10.1155/2023/4127486., articleTitle=Learning to drive in the NGSIM simulator using proximal policy optimization, refAbstract=null), Reference(id=1172170795020305019, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, doi=null, pmid=null, pmcid=null, year=2021, volume=23, issue=8, pageStart=10 333, pageEnd=10 342, url=null, language=null, rfNumber=[17], rfOrder=26, authorNames=CHEN Baiming, CHEN Xiang, WU Qiong, journalName=IEEE Transactions on Intelligent Transportation Systems, refType=null, unstructuredReference=CHEN Baiming, CHEN Xiang, WU Qiong, et al. Adversarial evaluation of autonomous vehicles in lane-change scenarios[J]. IEEE Transactions on Intelligent Transportation Systems, 2021, 23(8): 10 333-10 342., articleTitle=Adversarial evaluation of autonomous vehicles in lane-change scenarios, refAbstract=null)], funds=[Fund(id=1172170792646328925, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, awardId=51908054, language=CN, fundingSource=国家自然科学基金资助(51908054), fundOrder=null, country=null), Fund(id=1172170792726020702, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, awardId=2024JC-YBMS-301, language=CN, fundingSource=陕西省科技厅自然科学基础研究计划项目(2024JC-YBMS-301), fundOrder=null, country=null), Fund(id=1172170792784740959, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, awardId=2024BAA011, language=CN, fundingSource=湖北省技术创新计划科技重大项目(2024BAA011), fundOrder=null, country=null), Fund(id=1172170792877015648, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, awardId=ZDSYS202310, language=CN, fundingSource=纯电动汽车动力系统设计与测试湖北省重点实验室开放基金资助(ZDSYS202310), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1172170789265719842, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, xref=1, ext=[AuthorCompanyExt(id=1172170789274108451, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, companyId=1172170789265719842, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 School of Vehicle Engineering,Xi'an Aeronautical Institute,Xi'an Shaanxi 710077,China), AuthorCompanyExt(id=1172170789282497060, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, companyId=1172170789265719842, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 西安航空学院 车辆工程学院,陕西 西安 710077)]), AuthorCompany(id=1172170789349605925, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, xref=2, ext=[AuthorCompanyExt(id=1172170789357994534, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, companyId=1172170789349605925, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 Hubei Key Laboratory of Power System Design and Test for Electrical Vehicle,Hubei University of Arts and Science,Xiangyang Hubei 441053,China), AuthorCompanyExt(id=1172170789366383143, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, companyId=1172170789349605925, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 湖北文理学院 纯电动汽车动力系统设计与测试湖北省重点实验室,湖北 襄阳 441053)]), AuthorCompany(id=1172170789425103400, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, xref=3, ext=[AuthorCompanyExt(id=1172170789433492009, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, companyId=1172170789425103400, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 School of Automotive and Traffic Engineering,Hubei University of Arts and Science,Xiangyang Hubei 441053,China), AuthorCompanyExt(id=1172170789446074922, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, companyId=1172170789425103400, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 湖北文理学院 汽车与交通工程学院,湖北 襄阳 441053)])], figs=[ArticleFig(id=1172170791035716167, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, language=EN, label=Fig.1, caption=Vehicle cut-in environment based on SMARTS, figureFileSmall=VU590kSNGbvFNGVl5V91rg==, figureFileBig=wmhDPMxXtnwoQfJhSHjY4w==, tableContent=null), ArticleFig(id=1172170791123796552, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, language=CN, label=图1, caption=基于SMARTS的车辆切入环境, figureFileSmall=VU590kSNGbvFNGVl5V91rg==, figureFileBig=wmhDPMxXtnwoQfJhSHjY4w==, tableContent=null), ArticleFig(id=1172170791182516809, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, language=EN, label=Fig.2, caption=Extraction of adjacent vehicles and target point, figureFileSmall=FGT2LnC7kZNKimh8TijlMw==, figureFileBig=XD+v2+HFDfeJCKjc5f78gQ==, tableContent=null), ArticleFig(id=1172170791245431370, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, language=CN, label=图2, caption=附近车辆的提取与目标点, figureFileSmall=FGT2LnC7kZNKimh8TijlMw==, figureFileBig=XD+v2+HFDfeJCKjc5f78gQ==, tableContent=null), ArticleFig(id=1172170791299957323, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, language=EN, label=Fig.3, caption=Change in mean accumulated reward during training, figureFileSmall=c3l0gAqCs3OkyMw8r++A6w==, figureFileBig=NLpMrNyzZC2+Wx7fv9t6bw==, tableContent=null), ArticleFig(id=1172170791367066188, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, language=CN, label=图3, caption=训练过程中累积奖励平均值变化, figureFileSmall=c3l0gAqCs3OkyMw8r++A6w==, figureFileBig=NLpMrNyzZC2+Wx7fv9t6bw==, tableContent=null), ArticleFig(id=1172170791425786445, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, language=EN, label=Fig.4, caption=Change in success rate of ego vehicle's hazardous cut-in during training, figureFileSmall=4N0gpMx57qxul9DOAu2NWg==, figureFileBig=7bhxY4NR1SOwMICHuQuxdw==, tableContent=null), ArticleFig(id=1172170791480312398, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, language=CN, label=图4, caption=训练过程中自车危险切入成功率的变化, figureFileSmall=4N0gpMx57qxul9DOAu2NWg==, figureFileBig=7bhxY4NR1SOwMICHuQuxdw==, tableContent=null), ArticleFig(id=1172170791551615567, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, language=EN, label=Fig.5, caption=Test results of model in different traffic flow density scenarios, figureFileSmall=zTnCO8sHEjmcmWjCYion3g==, figureFileBig=BCr/3vJCMp2E6V+JRvaxag==, tableContent=null), ArticleFig(id=1172170791610335824, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, language=CN, label=图5, caption=模型在不同交通流密度场景中的测试结果, figureFileSmall=zTnCO8sHEjmcmWjCYion3g==, figureFileBig=BCr/3vJCMp2E6V+JRvaxag==, tableContent=null), ArticleFig(id=1172170791673250385, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, language=EN, label=Fig.6, caption=Distribution of time to collision between ego vehicle and target vehicle when ego vehicle cut in during model test, figureFileSmall=BHfYhcuSwaWkW+y19ItaWw==, figureFileBig=Bm2lFsqrGaGL9j/Cwalk2A==, tableContent=null), ArticleFig(id=1172170791740359250, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, language=CN, label=图6, caption=模型测试中自车在切入时刻与目标车的碰撞时间分布, figureFileSmall=BHfYhcuSwaWkW+y19ItaWw==, figureFileBig=Bm2lFsqrGaGL9j/Cwalk2A==, tableContent=null), ArticleFig(id=1172170791828439635, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, language=EN, label=Fig.7, caption=A test example, figureFileSmall=ShwbW37eBtqLuMlRiFzdUg==, figureFileBig=TFg0/ejmD0whTpOoK6SaXg==, tableContent=null), ArticleFig(id=1172170791895548500, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, language=CN, label=图7, caption=测试实例, figureFileSmall=ShwbW37eBtqLuMlRiFzdUg==, figureFileBig=TFg0/ejmD0whTpOoK6SaXg==, tableContent=null), ArticleFig(id=1172170791971045973, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, language=EN, label=Fig.8, caption=Change of output acceleration of tested intelligent driving strategy, figureFileSmall=2i6TckGB5b85U1M9VNKbzA==, figureFileBig=kgsiUDRfOc22G63IivPM3g==, tableContent=null), ArticleFig(id=1172170792033960534, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, language=CN, label=图8, caption=被测智能驾驶策略输出加速度变化, figureFileSmall=2i6TckGB5b85U1M9VNKbzA==, figureFileBig=kgsiUDRfOc22G63IivPM3g==, tableContent=null), ArticleFig(id=1172170792096875095, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, language=EN, label=Table 1, caption=

State representation

, figureFileSmall=null, figureFileBig=null, tableContent=
类型 特征描述
自车 纵向车速、横向车速、朝向、车道偏移距离
附近车辆(i为车辆数) 距离 d i、纵向距离 d i x、横向距离 d i y
目标点与目标车(T为目标车,m为目标点) 与目标点的距离为 d m,自车与目标车的纵向距离 d T x,横向距离 d T y,相对车速 Δ v o = d x T 2 + d y T 2 Δ v
), ArticleFig(id=1172170792193344088, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, language=CN, label=表1, caption=

状态表示

, figureFileSmall=null, figureFileBig=null, tableContent=
类型 特征描述
自车 纵向车速、横向车速、朝向、车道偏移距离
附近车辆(i为车辆数) 距离 d i、纵向距离 d i x、横向距离 d i y
目标点与目标车(T为目标车,m为目标点) 与目标点的距离为 d m,自车与目标车的纵向距离 d T x,横向距离 d T y,相对车速 Δ v o = d x T 2 + d y T 2 Δ v
), ArticleFig(id=1172170792281424473, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, language=EN, label=Table 2, caption=

Hyperparameters of TD3

, figureFileSmall=null, figureFileBig=null, tableContent=
参数 取值 参数 取值 参数 取值
学习率 3×10-4 经验回放
池大小
1×106 软更新率 0.005
批大小 100 折扣因子 0.99 总步数 5×105
), ArticleFig(id=1172170792356921946, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, language=CN, label=表2, caption=

TD3超参取值

, figureFileSmall=null, figureFileBig=null, tableContent=
参数 取值 参数 取值 参数 取值
学习率 3×10-4 经验回放
池大小
1×106 软更新率 0.005
批大小 100 折扣因子 0.99 总步数 5×105
), ArticleFig(id=1172170792432419419, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, language=EN, label=Table 3, caption=

Comparison of success rate of ego vehicle's hazardous cut-in among models trained with different algorithms

, figureFileSmall=null, figureFileBig=null, tableContent=
算法 危险切入成功率/%
TD3 80.35
DDPG 29.57
PPO 68.85
), ArticleFig(id=1172170792495333980, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106707026309858, language=CN, label=表3, caption=

不同算法所训练模型的危险切入成功率对比

, figureFileSmall=null, figureFileBig=null, tableContent=
算法 危险切入成功率/%
TD3 80.35
DDPG 29.57
PPO 68.85
)], attaches=null, journal=Journal(id=1123942128916217864, delFlag=0, nameCn=中国安全科学学报, nameEn=China Safety Science Journal, nameHistory1=null, nameHistory2=null, issn=1003-3033, eissn=, cn=11-2865/X, coden=null, periodic=0, language=CN, oaType=0, ccby=null, superviseOffice=null, ownerOffice=null, pubOffice=null, editorOffice=null, officeType=null, aims=null, clcCode=null, officeProv=null, officeCity=null, officeAddr=null, officeZip=null, officeEmail=null, officePhone=null, editDirector=null, officeDirector=null, officeDirectorPhone=null, officeStaffNum=null, officeEmpNum=null, coverPicUrl=fkqsFM6VKlHC4gCtS5XqTw==, journalPrice=null, startedYear=null, abbrevIsoEn=Chin Saf Sci J, journalRemark=null, publicationField=null, createdTime=null, updatedTime=1754269350027, createdBy=null, updatedBy=13701087609, firstLetterCn=C, firstLetterEn=C, subjectCode=Engineering, subjectName=工程, subjectCodeEn=Engineering, subjectNameEn=null, picCn=fkqsFM6VKlHC4gCtS5XqTw==, picEn=SHn9HgqSxtJrOcAxqD++4Q==, jcr=null, cjcr=null, exts=[JournalExt(id=1159052918994595848, language=CN, name=中国安全科学学报, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=null, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=http://www.cssjj.com.cn/, createdTime=1754269350050, updatedTime=1754269350050, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=http://www.cssjj.com.cn/CN/column/item15.shtml, submissionAuthorUrl=https://zgaqkxxbauthor.manuscriptcloud.com/login, submissionEditorUrl=https://zgaqkxxbeditor.manuscriptcloud.com/login, submissionReviewUrl=https://zgaqkxxbauthor.manuscriptcloud.com/login, submissionCeEditorUrl=https://zgaqkxxbeditor.manuscriptcloud.com/login, submissionAeEditorUrl=https://zgaqkxxbeditor.manuscriptcloud.com/login, option={"copyright":""}), JournalExt(id=1159052919040733193, language=EN, name=China Safety Science Journal, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=null, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=http://www.cssjj.com.cn/EN/1003-3033/home.shtml, createdTime=1754269350061, updatedTime=1754269350061, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=https://synbioj.cip.com.cn/EN/column/column3.shtml, submissionAuthorUrl=https://zgaqkxxbauthor.manuscriptcloud.com/login, submissionEditorUrl=https://zgaqkxxbeditor.manuscriptcloud.com/login, submissionReviewUrl=https://zgaqkxxbauthor.manuscriptcloud.com/login, submissionCeEditorUrl=https://zgaqkxxbeditor.manuscriptcloud.com/login, submissionAeEditorUrl=https://zgaqkxxbeditor.manuscriptcloud.com/login, option={"copyright":""})], databaseList=null, tenantJournalId=1146031787341344770, websiteList=[Website(id=1148243202345263519, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146031787341344770, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/zgaqkxxb/CN, language=CN, createTime=1751692112766, createBy=18614031015, updateTime=1753502583634, updateBy=18614031015, name=《中国安全科学学报》中文站点, tplId=1146099689490845704, title=中国安全科学学报, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1148618794941046792, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202345263519, code=articleTextType, value=kx, createTime=1751781661020, updateTime=1751781661020, creator=18614031015, updator=18614031015), WebsiteProps(id=1148618794911686661, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202345263519, code=banner, value=null, createTime=1751781661012, updateTime=1751781661012, creator=18614031015, updator=18614031015), WebsiteProps(id=1148618794894909444, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202345263519, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=tui0IVO9FMwB61HHtX5scg==, createTime=1751781661008, updateTime=1751781661008, creator=18614031015, updator=18614031015), WebsiteProps(id=1148618794932658183, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202345263519, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1751781661017, updateTime=1751781661017, creator=18614031015, updator=18614031015), WebsiteProps(id=1148618794924269574, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202345263519, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1751781661015, updateTime=1751781661015, creator=18614031015, updator=18614031015)]), Website(id=1155836763751993353, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146031787341344770, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/zgaqkxxb/EN, language=EN, createTime=1753502558893, createBy=18614031015, updateTime=1753524450387, updateBy=18614031015, name=《中国安全科学学报》英文站点, tplId=1146101810881728533, title=China Safety Science Journal, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1155895925743669425, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155836763751993353, code=articleTextType, value=kx, createTime=1753516664205, updateTime=1753516664205, creator=18614031015, updator=18614031015), WebsiteProps(id=1155895925722697902, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155836763751993353, code=banner, value=null, createTime=1753516664200, updateTime=1753516664200, creator=18614031015, updator=18614031015), WebsiteProps(id=1155895925714309293, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155836763751993353, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=tui0IVO9FMwB61HHtX5scg==, createTime=1753516664198, updateTime=1753516664198, creator=18614031015, updator=18614031015), WebsiteProps(id=1155895925735280816, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155836763751993353, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1753516664203, updateTime=1753516664203, creator=18614031015, updator=18614031015), WebsiteProps(id=1155895925731086511, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155836763751993353, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1753516664202, updateTime=1753516664202, creator=18614031015, updator=18614031015)])], journalTitle=中国安全科学学报, weixinUrl=null, journalUrl=null, iacademicId=null, status=0, seqNo=null, journalTitleEn=China Safety Science Journal, journalPhotoCn=fkqsFM6VKlHC4gCtS5XqTw==, journalPhotoEn=SHn9HgqSxtJrOcAxqD++4Q==, journalFirstLetter=C, journalRecommend=null, journalNew=null, journalCollection=1, jcrJf=null, cjcrJf=null, jcrJfStr=null, cjcrJfStr=null, submissionFirstDecision=null, sciSubjectClassification=null, casSubjectClassification=null, citeScore=null, totalCitationFrequency=null, icpCode=null, psCode=null, advertisingLicenseCode=null, copyrightInformation=null, country=null, option=null, provinceCode=null, provinceName=null, collectFlag=false), detailUrlCn=https://castjournals.cast.org.cn/joweb/zgaqkxxb/CN/10.16265/j.cnki.issn1003-3033.2025.01.0540, detailUrlEn=https://castjournals.cast.org.cn/joweb/zgaqkxxb/EN/10.16265/j.cnki.issn1003-3033.2025.01.0540, pdfUrlCn=https://castjournals.cast.org.cn/joweb/zgaqkxxb/CN/PDF/10.16265/j.cnki.issn1003-3033.2025.01.0540, pdfUrlEn=https://castjournals.cast.org.cn/joweb/zgaqkxxb/EN/PDF/10.16265/j.cnki.issn1003-3033.2025.01.0540, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
用于自动驾驶测试的车辆危险切入策略研究
收藏切换
PDF下载
周扬 1, 2 , 陈运星 2, 3, ** , 吴玲 1
中国安全科学学报 | 安全工程技术 2025,35(1): 112-119
收起
收藏切换
中国安全科学学报 | 安全工程技术 2025, 35(1): 112-119
用于自动驾驶测试的车辆危险切入策略研究
全屏
周扬1, 2 , 陈运星2, 3, ** , 吴玲1
作者信息
  • 1 西安航空学院 车辆工程学院,陕西 西安 710077
  • 2 湖北文理学院 纯电动汽车动力系统设计与测试湖北省重点实验室,湖北 襄阳 441053
  • 3 湖北文理学院 汽车与交通工程学院,湖北 襄阳 441053
  • 周扬 (1989—),男,陕西汉中人,博士,副教授,主要从事人-车-路系统安全、自动驾驶测试等方面的研究。E-mail:

    陈运星 副教授

    吴玲 副教授

通讯作者:

**陈运星(1987—),男,湖北荆门人,博士,副教授,主要从事驾驶行为感知、智能驾驶技术等方面的研究。E-mail:
Research on vehicle hazardous cut-in strategy used in autonomous driving test
Yang ZHOU1, 2 , Yunxing CHEN2, 3, ** , Ling WU1
Affiliations
  • 1 School of Vehicle Engineering,Xi'an Aeronautical Institute,Xi'an Shaanxi 710077,China
  • 2 Hubei Key Laboratory of Power System Design and Test for Electrical Vehicle,Hubei University of Arts and Science,Xiangyang Hubei 441053,China
  • 3 School of Automotive and Traffic Engineering,Hubei University of Arts and Science,Xiangyang Hubei 441053,China
出版时间: 2025-01-28 doi: 10.16265/j.cnki.issn1003-3033.2025.01.0540
文章导航
收藏切换

为提高车辆切入测试场景中交通车的交互能力,提出一种基于深度强化学习方法的车辆危险切入策略设计方法。首先,基于可扩展多智能体强化学习培训学校(SMARTS)仿真平台构建仿真环境;然后,采用双延迟深度确定性策略梯度算法(TD3)训练智能体危险切入随机选定的目标车辆,将该算法与近端策略优化算法(PPO)和深度确定性策略梯度算法(DDPG)进行对比,在7种不同车辆密度的场景中测试训练后的模型;最后,构建多智能体测试环境,将所训练模型用于智能驾驶策略的验证。结果表明:模型在训练中的危险切入成功率达80.35%,优于2种对比方法;在模型测试中,除2 700辆/h测试场景外,该模型在另外3个未在训练中使用的测试场景均达到80%以上的危险切入成功率,显示出良好的泛化能力。同时,切入时刻与目标车的碰撞时间值显示95%集中在0~6 s,取值在(0,2]、(2,4]和(4,6]s的占比分别为60%、30%和5%,可覆盖具有不同碰撞风险的测试工况。在智能驾驶策略验证中,采用所训练模型控制的交通车能主动切入至待测车辆前方,使待测车辆面临追尾风险,有助于发现智能驾驶策略的安全隐患。

自动驾驶  /  车辆危险切入  /  虚拟测试  /  危险场景  /  强化学习

To improve the interaction ability of traffic vehicles in the cut-in scenario,a method for constructing a vehicle hazardous cut-in strategy based on deep reinforcement learning was proposed. Firstly,a simulated environment was built based on scalable multi-agent reinforcement learning training school(SMARTS) simulation platform. Then,twin delayed deep deterministic policy gradients (TD3) algorithm was adopted to train an agent to cut in a randomly chosen target vehicle hazardously. The algorithm was compared with proximal policy optimization (PPO) and deep deterministic policy gradient (DDPG) algorithms. The trained model was tested in seven different scenarios with varying traffic densities. Finally,a multi-agent testing environment was built,and the trained model was applied to validate intelligent driving strategies. The results show that the success rate of hazardous cut-ins reaches 80.35% in model training with TD3 algorithm,outperforming both comparative methods. In model testing,except for the 2 700 vehicle/h test scenario,the model achieves a hazardous cut-in success rate of over 80% in the other three test scenarios that were not used in training,demonstrating good generalization ability. Meanwhile,the time to collision values between the ego vehicle and the target vehicle at the moment of lane changes are concentrated within the range of 0 to 6 seconds,with 95% falling within this bracket. The proportions of time to collision values in the intervals of (0,2],(2,4],(4,6]s are 60%,30%,and 5% respectively,covering test conditions with different collision risk. In the validation of intelligent driving strategies,the traffic vehicle controlled by the trained model can actively perform cut-ins in front of the test vehicles,exposing it to the risk of a rear-end collision and helping in identifying safety vulnerabilities in intelligent driving strategies.

autonomous driving  /  vehicle hazardous cut-in  /  virtual tests  /  hazardous scenarios  /  reinforcement learning
周扬, 陈运星, 吴玲. 用于自动驾驶测试的车辆危险切入策略研究. 中国安全科学学报, 2025 , 35 (1) : 112 -119 . DOI: 10.16265/j.cnki.issn1003-3033.2025.01.0540
Yang ZHOU, Yunxing CHEN, Ling WU. Research on vehicle hazardous cut-in strategy used in autonomous driving test[J]. China Safety Science Journal, 2025 , 35 (1) : 112 -119 . DOI: 10.16265/j.cnki.issn1003-3033.2025.01.0540
自动驾驶汽车(Autonomous Vehicles,AVs)为解决传统汽车存在的交通拥堵、事故频发和提升驾驶舒适度提供了新的解决方案,成为近年来的研究焦点[1]。在实际部署AVs前,为确保其安全性而进行以安全性为主的性能测试至关重要。在开放道路环境下测试是AVs测试的常用方法,由于开放道路环境对AVs安全性产生挑战的危险场景极为罕见,因此,AVs需行驶数十亿英里才能证明其安全性是否低于人类驾驶。从测试成本、效率以及安全性等多方面考虑,开放道路测试已无法满足AVs技术研发和迭代的要求[2]。虚拟测试具有场景配置灵活、测试成本低及重复性强的优势,在虚拟条件下可构建开放道路中小概率发生的危险场景,已成为AVs测试的主要方式[3]。在虚拟测试研究中,基于危险场景的AVs虚拟测试有利于缩短测试周期,实现对AVs的加速测试。因此,构建对AVs具有挑战的危险测试场景已成为重要的研究内容。
在相关研究中,郭柏仓等[4]提取了59例实车变道切入数据片段,采用危险感知系数表征场景的危险程度,分析提取了与场景危险程度显著相关的场景要素,基于K-means算法聚类得到4类城市道路车辆变道切入场景。赵祥模等[5]提出了一种紧急变道轨迹对抗生成方法,基于公开数据集中的紧急变道数据进行训练,构建了包含5万条变道轨迹的车辆危险变道测试场景库。朱冰等[6]提出了一种基于危险边界搜索的AVs加速测试方法,以车辆切入场景为例,采用所提出方法可得到所有碰撞场景的危险边界,使测试时长缩短73%。为使所构建的危险测试场景中交通车具备交互能力,相关研究开始采用机器学习方法构建具有连续决策能力的交通车模型,生成危险测试场景测试AVs的连续决策能力。其中,FENG Shuo等[7]设计了用于自动驾驶虚拟测试的自然-对抗驾驶环境(Naturalistic and Adversarial Driving Environment,NADE),在构建的自然驾驶环境训练场景中交通车在特定时刻执行特定动作,生成自然而有挑战的测试环境,有效降低AVs的测试里程。FENG Shuo等[8]设计了自演绎的AVs测试环境,环境中的背景车辆采用提出的密集深度强化学习算法训练,该算法保留了数据中的安全关键事件信息,起到了AVs安全性加速测试的作用。SUN Haowei等[9]提出了一种用于生成自动驾驶测试边界用例的统一框架,采用深度强化学习方法构建场景中背景车辆的行驶策略,背景车辆采取攻击性驾驶风格与被测车辆进行互动,以产生更多的边界测试场景。李江坤等[10]提出了一种AVs加速测试的危险场景强化生成方法,利用神经网络来建立场景的动力学系统模型,通过强化学习算法构建场景控制器,生成了具有交互博弈的危险测试场景。以上研究主要基于采样的方法生成车辆切入危险测试场景,场景中交通车采用预设的行驶轨迹,存在交通车交互能力不足的缺陷。为使交通车具备交互能力,相关研究开始采用强化学习方法构建交通车控制策略,然而,目前对于在车辆切入场景中构建兼具危险切入及交互能力的交通车控制策略的研究仍较为缺乏。
鉴于此,笔者拟采用深度强化学习方法构建具备交互能力的车辆危险切入策略,通过可扩展多智能体强化学习培训学校平台建立车辆切入仿真环境,基于双延迟深度确定性策略梯度(Twin Delayed deep deterministic policy gradients,TD3)算法构建车辆危险切入模型,使用训练后的模型测试被测系统的安全性,以期为AVs危险测试场景中交通车控制模型的构建提供参考。
建立具有较高逼真度的仿真环境是模型训练的前提,提高仿真环境的多样性是确保模型能适应环境变化的基础。
选择基于SMARTS平台[11]构建车辆切入仿真环境。SMARTS通过PyBullet物理引擎提供真实的车辆运动仿真,基于开源交通仿真软件——城市交通能力仿真(Simulation of Urban Mobility,SUMO)来生成仿真道路和交通流,SMARTS还提供基于网页的三维可视化,方便用户查看仿真结果。构建如图1所示的车辆切入环境,车道长度为200m,场景中背景车辆车速通过SUMO控制。具体而言,背景车辆会综合车道限速值、最高期望车速及SUMO内置的跟车模型所输出车速值,选择其中的最低值作为目标车速。自车由强化学习智能体控制。
交通流是场景的核心要素之一[12],为增加场景中交通流的多样性,采取如下措施:①参考胡祥旺等[13]的研究,为更加接近贴近真实交通环境中不断变化的交通流密度,生成3种不同流量水平的训练场景,分别对应低、中、高流量。设定这3种场景在单车道上的车流量分别为1 200、1 800和2 400辆/h,随机生成场景中交通车的初始间距、车速及位置,3种场景中共包含1 200种不同初始化的交通流。②在智能体与环境交互的每一回合开始阶段,随机选择交通流中的任意一辆交通车作为目标车;自车从起始位置出发,驶入目标车所在车道并位于目标车辆前方时,认定自车完成切入。根据周文帅等[14]的研究,在车辆切入场景中,若自车在切入时刻与目标车的碰撞时间为0~6s,则定义为高风险切入场景。因此,当自车成功切入目标车前方,且切入时刻的碰撞时间值为0~6s,则认为自车危险切入成功。
强化学习算法通过智能体与环境互动以及接收环境反馈来进行试错学习,从而赋予智能体与环境的交互能力。通过设计特定的奖励函数,可引导智能体展现出期望的行为。因此,为构建同时具备危险切入及交互能力的交通车控制策略,选择深度强化学习算法中代表方法TD3来构建车辆危险切入策略。
1) TD3算法。TD3算法作为深度确定性策略梯度算法(Deep Deterministic Policy Gradient,DDPG)算法的改良版本,包括双Q网络、延迟策略更新和目标策略平滑的改进措施,有效改善了DDPG算法的性能和稳定性[15]。为加强智能体对环境的探索,对TD3模型的主策略网络所输出动作 a t添加从正态分布中采样的噪声,见下式:
a t = c l i p [ π ω ( s t ) + N ( μ σ ) a m i n a m a x ]
式中: t为时刻; π ω为TD3策略模型, ω为参数; N ( μ σ )表示均值 μ、方差 σ的正态分布,二者取值分别为0和0.1; s t t时刻的状态; a m i n a m a x分别为动作取值的下限和上限;clip函数截取了添加噪声后的动作,使其处于[ a m i n a m a x ]的区间。
2) 状态表示和动作空间。所采用的状态表示由3部分构成,见表1。第1部分为自车特征,共4项;第2部分是附近车辆特征,为提取与自车决策最相关的信息,过滤SMARTS所提供的周围车辆信息,仅提取与自车距离最近的6辆车辆(图2中除自车外的其他车辆),共计18项特征;第3部分是目标车与目标点特征,由于与自车距离最近的6辆附近车辆中可能并不包含目标车,因此,需额外提取目标车辆信息,目标点(用星形标志表示)是指目标车所在车道正前方10m的一个预设点。
由于o的计算结果取值范围较广,为减小模型训练中参数的搜索空间,将o计算值限制在(0,20]s范围。采用连续动作空间,TD3模型输出每个时刻车辆的加速度和横摆角速度值。为将模型的输出动作值限制在合理范围,将加速度范围设为[-5,5]m/s2,将横摆角速度范围设为[-0.5,0.5]rad/s,当动作值超过设定区间的上限或下限值时,则取对应的上、下限值。
3) 奖励函数。奖励函数是强化学习的核心,设计合适的奖励可加快模型收敛并提升模型的性能表现。所设计奖励函数包括危险切入即时奖励、引导自车危险切入奖励和不当行为惩罚3部分。
危险切入即时奖励。根据1.1节中所定义的危险切入判定标准,记自车切入目标车所在车道且位于目标车前方时刻为tc,若该时刻自车与目标车的碰撞时间 o t c∈(0,6]s,则给予智能体较大奖励;若 o t c未落入上述区间,则依据自车切入时刻与目标车的相对车速Δ v t c确定奖励值,Δ v t c越大,目标车追尾自车的风险则更高,奖励值相应增加,危险切入即时奖励值 r d c见下式:
r d c = 1   000 o t c 6 10 + Δ v t c
引导自车危险切入奖励。为引导智能体控制自车行驶至目标车前方,在目标车所在车道正前方距离10m处设置一个虚拟的目标点,根据自车与该目标点的距离 d m设计如下的距离奖励值 r d 随着 d m减小, r d增大,即自车与目标点距离越近,值越高。因此, r d鼓励智能体控制自车行驶至该目标点,到达目标车前方。
r d = 1 - d m n 1 0.4
式中 n 1为对 d m进行归一化所采用的系数,取50。
由于强化学习训练是以最大化累积奖励为目标,仅设置距离奖励会导致智能体以尽可能快的车速行驶至目标车前方以收集更多奖励,导致自车在切入至目标车前方时容易出现车速大于目标车车速的情况,不能满足文中所定义危险切入的要求。为解决该问题,定义碰撞时间折扣系数D,当自车与目标点的距离小于5m时,按照下式计算其值,可以看出,o值越小则D越大。当自车与目标点较远时,将该值定为1。
D = 1 - o n 2 0.4
式中 n 2为对o进行归一化的系数,取20。若o为负值,则取20代入式(4)计算。
最后,通过下式计算得到引导自车危险切入奖励项 r y d
r y d = r d · D
不当行为惩罚。为防止自车在行驶过程中出现如驶出道路、倒车或反向行驶、停止不动及主动与交通车碰撞4种不当行为,将这些行为的奖励值设为-10,作为不当行为惩罚项 r p。由于自车在危险切入时很容易造成目标车追尾,为不影响智能体对环境的探索,当自车被交通车追尾时不予惩罚。
所采用的奖励函数为上述3部分奖励的线性叠加,见下式:
R = θ 1 · r d c + θ 2 · r y d + θ 3 · r p
式中 θ 1 θ 2 θ 3 r d c r y d r p等3种奖励对应的权重系数。
参照原算法,所构建TD3模型中的6个神经网络模型均采用2层全连接神经网络结构,神经元数量分别为400和300,模型训练所采用的超参取值见表2,其余参数设置均与原算法保持一致。
由于TD3算法是针对DDPG算法的改进,故将DDPG算法作为比较方法。DDPG模型中,神经网络的结构及神经元数量、模型训练所采用超参取值均与TD3算法保持一致。
将基于策略梯度的深度强化学习代表算法近端策略优化算法(Proximal Policy Optimization,PPO)作为比较方法,PPO算法通过限制策略更新的幅度来稳定训练过程。PPO模型中的Actor和Critic网络结构与TD3保持一致,模型训练所采用的学习率和总训练步数与TD3的设置相同,其他超参的选择参照了ZHOU Yang等[16]的研究。
为确定奖励权重系数 θ 1 θ 2 θ 3的最佳取值,在3种奖励权重系数的不同取值下训练模型。在训练过程中,统计每个回合中自车完成危险切入的次数,并将成功危险切入的回合数除以记录间隔内的总回合数,以此得到模型在训练过程中的危险切入成功率,作为模型性能的评价指标。在选取8种不同奖励权重系数组合训练后发现,3种奖励权重系数的取值均为1/3时,危险切入成功率达到了较高水平。由于奖励值的大小不会对模型性能产生影响,为便于计算,将3种奖励权重系数的取值均设为1。
1) 模型训练结果。在训练过程中记录了智能体在每个回合累积奖励平均值的变化,结果如图3所示。可以发现,在训练过程中,TD3智能体每回合累积的平均奖励值呈上升趋势。当累积步数达到3.5×105步时,累积奖励的平均值趋于平稳,表明TD3模型已经收敛。
图4为TD3模型在训练过程中危险切入成功率的变化,为说明本文设计的奖励函数对智能体完成危险切入任务的影响,将TD3算法在其他2种奖励函数设置下的训练结果作为对比。图中3种奖励函数 R 1 R 2 R 3的区别仅在于第2项,其中,奖励 R 1采用式(6)的完整奖励函数,奖励 R 2相比 R 1在奖励组成的第2部分仅采用距离奖励值rd,未采用式(5)中的TTC折扣系数缩减距离奖励,奖励 R 3采用SMARTS内置的代表智能体每步行驶距离的 r e作为奖励的第2部分。
从训练结果看出,TD3模型在训练过程中最高可达到80%左右的危险切入成功率。对比TD3算法在3种不同奖励函数设置下的训练结果,可以发现,相比采用本文所提出的奖励函数 R 1,采用奖励 R 2时,虽然智能体仍能学习并不断提升,但最终仅能达到65%左右的危险切入成功率,而当采用奖励 R 3时,智能体的危险切入成功率一直为0,说明智能体不能学到危险切入行为。上述结果表明:文中所设计的奖励函数有利于智能体探索环境中的高奖励状态区域,使智能体能够尽快掌握本文所定义的危险切入行为,最终也取得了更好的性能表现。使用相同的奖励函数对DDPG和PPO模型进行了训练,将不同模型在收敛状态下的危险切入成功率结果列于表3中。结果显示,采用TD3算法所训练模型取得了最高的80.35%的危险切入成功率,采用PPO算法所训练的模型的危险切入成功率略低于TD3模型。与二者相比,DDPG模型仅能达到29.57%的危险切入成功率。由于TD3算法在DDPG算法的基础上进行了多方面的改进,在文中所定义的车辆危险切入任务中,TD3算法的表现显著优于DDPG算法。
2) 模型测试结果。为了说明模型的泛化能力,将训练得到的最优模型在包括900、1 200、1 500、1 800、2 100、2 400、2 700辆/h的7种不同车辆密度场景中进行测试,每种场景均测试100个回合。
图5为模型在7种车辆密度场景中危险切入成功率的对比,其中,在包括900~2 400辆/h的6种场景中,模型均能达到80%以上的危险切入成功率,仅在车辆密度达到2 700辆/h后,危险切入成功率下降至63%。随着场景车辆密度提高,自车完成危险切入任务的难度也随之提高,危险切入成功率随之下降,尤其是在车辆密度达到2 700辆/h后,经常出现目标车与相邻车道车辆并排行驶的情况,自车无法行驶至目标车前方。由于模型训练仅采用了1 200、1 800、2 400辆/h 3种场景,对于其余4种训练中未使用的测试场景,除2 700辆/h场景中的危险切入成功率较低外,在其余场景中的危险切入成功率均与训练结果相似,说明所训练模型具有较好的泛化能力。
图6为模型在测试中的 o t c分布,自车在危险切入时该值有约95%比例的取值在(0,6]s内,其中,(0,2] s占60%,(2,4]s占30%,(4,6]s占5%,结果表明:所建模型能在切入过程中对目标车产生不同程度的风险,从而覆盖具有不同风险的测试工况。
3) 智能驾驶策略测试。为将所构建的车辆危险切入策略模型用于AVs安全性能测试,在所建立3车道环境的基础上构建了包含2个智能体的多智能体测试环境,其中,一个智能体采用待测智能驾驶策略控制待测车辆,另一智能体采用训练得到的车辆危险切入策略控制一辆测试车,以待测车辆作为目标车。环境中的其他交通车由SMARTS内置交通车模型来控制。
为验证该环境的测试效果,选用智能驾驶员模型(Intelligent Driving Model,IDM)作为被测智能驾驶策略。IDM模型是典型的跟车模型,也被其他研究选作被测系统的控制策略[6-7],其计算方法见下式。为防止IDM模型输出加速度值超过合理范围,将IDM模型输出的加速度值限制在[-7,7]m/s2范围。IDM模型参数设置参考CHEN Baiming等[17]的研究。
$a(t)=\alpha_{\max }\left[1-\left(\frac{v(t)}{\tilde{v}}\right)^{\beta}-\left(\frac{\tilde{h}(t)}{h(t)}\right)^{2}\right]$
$\tilde{h}(t)=h_{0}+v(t) \tilde{T}+\frac{v(t) \Delta v(t)}{2 \sqrt{a_{\max } a_{\mathrm{com}}}}$
式中: a h为自车的加速度和与前车间距,单位分别为m/s2和m; v Δ v为自车车速及自车与前车的相对车速,m/s; a m a x为自车最大加速度或减速度,取1 m/s2 a c o m指舒适减速度,取1.67 m/s2 v ~为期望速度,取10m/s; T ~为期望车头时距,取1.5s; h 0为阻塞间距,取2m; β为加速度指数,取4。
在所构建的多智能体测试场景中测试被测智能驾驶策略,采用SMARTS支持的Visdom可视化工具对测试过程进行可视化,将摄像头固定于被测车辆(最左侧虚线框)。图7为测试过程中的一个片段,可以发现,在仿真初始时刻(图7a),采用车辆危险切入策略所控制的测试车(中间车道实线框)位于被测车辆后方的中间车道,2车的纵向初始间距为6m;测试车在超过被测车辆一定距离后(图7b),开始主动降低车速,以确保在切入过程中对被测车辆产生碰撞风险;测试车于t=4.9s开始换道(图7c),最终于t=5.5s左右切入被测车辆前方(图7d),尽管被测车辆在t=5.5s迅速采取制动措施,制动减速度达到所设置减速度的上限-7m/s2,如图8所示。然而,被测车辆最终未能避免与测试车产生碰撞。测试结果表明:采用所提方法可生成对AVs安全性具有挑战的车辆切入危险测试场景,有助于发现AVs控制策略的安全隐患。
1) 与DDPG和PPO这2种比较方法相比,TD3算法在训练过程中能达到更高的危险切入成功率。将所训练模型在7种不同车辆密度场景中测试,除车流密度较高的2 700辆/h场景外,在其余场景均取得较高的危险切入成功率,说明模型泛化性能良好。模型所控制自车在切入时刻与目标车的碰撞时间值主要集中在(0,6]s范围内,能全面涵盖不同碰撞风险的测试工况。
2) 将所训练模型用于智能驾驶策略的验证,模型所控制交通车能主动危险切入至待测车前方,使待测车辆产生追尾风险,能够起到对待测智能驾驶策略的安全性进行验证的效果。
3) 在后续研究中,将考虑把训练得到的车辆危险切入策略模型集成到自动驾驶虚拟测试平台中,生成前车危险切入场景,以便测试自动驾驶策略。将考虑在包括弯道、十字路口等的复杂场景,以及基于自然驾驶数据的真实交通流中训练模型,提高模型对多样化环境的适应性。
  • 国家自然科学基金资助(51908054)
  • 陕西省科技厅自然科学基础研究计划项目(2024JC-YBMS-301)
  • 湖北省技术创新计划科技重大项目(2024BAA011)
  • 纯电动汽车动力系统设计与测试湖北省重点实验室开放基金资助(ZDSYS202310)
参考文献 引证文献
排序方式:
[1]
朱冰, 张培兴, 赵健, 等. 基于场景的自动驾驶汽车虚拟测试研究进展[J]. 中国公路学报, 2019, 32(6): 1-19.
ZHU Bing, ZHANG Peixing, ZHAO Jian, et al. Review of scenario-based virtual validation methods for automated vehicle[J]. China Journal of Highway and Transport, 2019, 32(6): 1-19.
[2]
邓伟文, 李江坤, 任秉韬, 等. 面向自动驾驶的仿真场景自动生成方法综述[J]. 中国公路学报, 2022, 35(1): 316-333.
DENG Weiwen, LI Jiangkun, REN Bingtao, et al. A survey on automatic simulation scenario generation methods for autonomous driving[J]. China Journal of Highway and Transport, 2022, 35(1): 316-333.
[3]
徐向阳, 胡文浩, 董红磊, 等. 自动驾驶汽车测试场景构建关键技术综述[J]. 汽车工程, 2021, 43(4): 610-619.
XU Xiangyang, HU Wenhao, DONG Honglei, et al. Review of key technologies for autonomous vehicle test scenario construction[J]. Automotive Engineering, 2021, 43(4): 610-619.
[4]
郭柏苍, 雒国凤, 金立生, 等. 面向自动驾驶虚拟测试的变道切入场景库构建方法[J]. 吉林大学学报:工学版, 2023, 53(11): 3130-3140.
GUO Baicang, LUO Guofeng, JIN Lisheng, et al. Construction method of cut-in scenario library for automatic driving virtual tests[J]. Journal of Jilin University: Engineering and Technology Edition, 2023, 53(11): 3130-3140.
[5]
赵祥模, 赵玉钰, 景首才, 等. 面向自动驾驶测试的危险变道场景泛化生成[J]. 自动化学报, 2023, 49 (10): 2211-2223.
ZHAO Xiangmo, ZHAO Yuyu, JING Shoucai, et al. Generalization generation of hazardous lane-changing scenarios for automated vehicle testing[J]. Acta Automatica Sinica, 2023, 49 (10): 2211-2223.
[6]
朱冰, 范天昕, 赵健, 等. 基于危险边界搜索的自动驾驶系统加速测试方法[J]. 吉林大学学报:工学版, 2023, 53(3): 704-712.
ZHU Bing, FAN Tianxin, ZHAO Jian, et al. Accelerate test method of automated driving system based on hazardous boundary search[J]. Journal of Jilin University: Engineering and Technology Edition, 2023, 53(3): 704-712.
[7]
FENG Shuo, YAN Xintao, SUN Haowei, et al. Intelligent driving intelligence test for autonomous vehicles with naturalistic and adversarial environment[J]. Nature Communications, 2021, 12(1): DOI: 10.1038/s41467-021-21007-8.
[8]
FENG Shuo, SUN Haowei, YAN Xintao, et al. Dense reinforcement learning for safety validation of autonomous vehicles[J]. Nature, 2023, 615(7953): 620-627.
[9]
SUN Haowei, FENG Shuo, YAN Xintao, et al. Corner case generation and analysis for safety assessment of autonomous vehicles[J]. Transportation Research Record, 2021, 2675(11): 587-600.
[10]
李江坤, 邓伟文, 任秉韬, 等. 基于场景动力学和强化学习的自动驾驶边缘测试场景生成方法[J]. 汽车工程, 2022, 44 (7): 976-986.
LI Jiangkun, DENG Weiwen, REN Bingtao, et al. Automatic driving edge scene generation method based on scene dynamics and reinforcement learning[J]. Automotive Engineering, 2022, 44(7): 976-986.
[11]
ZHOU Ming, LUO Jun, VILLELLA J, et al. Smarts: an open-source scalable multi-agent reinforcement training school for autonomous driving[C]. Conference on Robot Learning, 2021: 264-285.
[12]
罗崎瑞, 张道文, 周华, 等. 面向智能汽车预期功能安全的驾驶场景评价[J]. 中国安全科学学报, 2022, 32(8):140-145.
LUO Qirui, ZHANG Daowen, ZHOU Hua, et al. Evaluation on driving scenarios for safety of intended functionality of intelligent vehicles[J]. China Safety Science Journal, 2022, 32(8):140-145.
[13]
胡祥旺, 倪颖, 孙剑. 车联网环境下匝道汇入区瓶颈换道优化[J]. 同济大学学报:自然科学版, 2023, 51(9):1424-1432.
HU Xiangwang, NI Ying, SUN Jian. Freeway merging area lane changing advisory optimization under connected vehicles environment[J]. Journal of Tongji University: Natural Science, 2023, 51(9): 1424-1432.
[14]
周文帅, 朱宇, 赵祥模, 等. 面向高速公路车辆切入场景的自动驾驶测试用例生成方法[J]. 汽车技术, 2021(1): 11-18.
ZHOU Wenshuai, ZHU Yu, ZHAO Xiangmo, et al. Vehicle cut-in test case generation methods for testing of autonomous driving on highway[J]. Automobile Technology, 2021(1): 11-18.
[15]
FUJIMOTO S, HOOF H, MEGER D. Addressing function approximation error in actor-critic methods[C]. International Conference on Machine Learning (ICML), 2018: 1587-1596.
[16]
ZHOU Yang, CHEN Yunxing. Learning to drive in the NGSIM simulator using proximal policy optimization[J]. Journal of Advanced Transportation, 2023, 2023: DOI: 10.1155/2023/4127486.
[17]
CHEN Baiming, CHEN Xiang, WU Qiong, et al. Adversarial evaluation of autonomous vehicles in lane-change scenarios[J]. IEEE Transactions on Intelligent Transportation Systems, 2021, 23(8): 10 333-10 342.
2025年第35卷第1期
PDF下载
458
193
引用本文
BibTeX
文章信息
doi: 10.16265/j.cnki.issn1003-3033.2025.01.0540
  • 接收时间:2024-08-11
  • 首发时间:2025-07-05
  • 出版时间:2025-01-28
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2024-08-11
  • 修回日期:2024-10-20
基金
国家自然科学基金资助(51908054)
陕西省科技厅自然科学基础研究计划项目(2024JC-YBMS-301)
湖北省技术创新计划科技重大项目(2024BAA011)
纯电动汽车动力系统设计与测试湖北省重点实验室开放基金资助(ZDSYS202310)
作者信息
    1 西安航空学院 车辆工程学院,陕西 西安 710077
    2 湖北文理学院 纯电动汽车动力系统设计与测试湖北省重点实验室,湖北 襄阳 441053
    3 湖北文理学院 汽车与交通工程学院,湖北 襄阳 441053

通讯作者:

**陈运星(1987—),男,湖北荆门人,博士,副教授,主要从事驾驶行为感知、智能驾驶技术等方面的研究。E-mail:
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/zgaqkxxb/CN/10.16265/j.cnki.issn1003-3033.2025.01.0540
分享至
全文二维码

扫描看全文

引用本文
BibTeX
本文的引用情况
2种不同金属材料的力学参数

Family
属数
Number of
genus
种数
Number of
species
占总种数比例
Percentage of
total species (%)

Genus
种数
Number of
species
占总种数比例
Percentage of total
species (%)
鹅膏菌科Amanitaceae 2 11 5.26 鹅膏菌属 Amanita 10 4.78
小菇科 Mycenaceae 2 12 5.74 丝盖伞属 Inocybe 5 2.39
多孔菌科 Polyporaceae 8 14 6.70 蜡蘑属 Laccaria 5 2.39
红菇科 Russulaceae 3 23 11.00 小皮伞属 Marasmius 6 2.87
小菇属 Mycena 11 5.26
光柄菇属 Pluteus 5 2.39
红菇属 Russula 17 8.13
栓菌属 Trametes 5 2.39
关闭全屏